company-research
Scannednpx machina-cli add skill blisspixel/primr/company-research --openclawFiles (1)
SKILL.md
2.5 KB
Company Research
Workflow
- Check prior research:
get_hypotheses(company)-- present any existing findings to user - Estimate cost:
estimate_run(company, url, mode)-- always before starting - Get user approval: show estimated cost and duration
- Start research:
research_company(company, url, mode)-- returns job_id immediately - Monitor: poll
check_jobs()every 2 minutes until complete - Present results: show report path and any new hypotheses
Research Modes
See references/modes.md for mode details (cost, duration, use cases).
Hypothesis Management
When user confirms or rejects a claim during research:
save_hypothesis(company, hypothesis_id, "validated", evidence)
save_hypothesis(company, hypothesis_id, "invalidated", evidence)
Error Handling
| Error | Resolution |
|---|---|
budget_exceeded | Hook blocked operation; request budget increase |
ssrf_blocked | URL failed security check; use deep mode instead |
qa_below_threshold | Report quality low; suggest refinement |
job_already_running | Wait for current job; use check_jobs |
Example Workflow
User: "Research Acme Corp at https://acme.com"
1. get_hypotheses("Acme Corp")
-> Found 2 prior hypotheses from last session
2. Present to user:
"I found prior research on Acme Corp:
- [VALIDATED] Uses microservices architecture
- [UNTESTED] Revenue growth exceeds 20% YoY
Shall I continue with new research?"
3. estimate_run("Acme Corp", "https://acme.com", "full")
-> Cost: $3.50, Time: ~30 minutes
4. Request approval:
"Full research will cost ~$3.50 and take ~30 minutes.
Reply 'approve' to proceed."
5. research_company("Acme Corp", "https://acme.com", "full")
-> Job started: job_abc123
6. Poll check_jobs() until complete
7. Present results and new hypotheses
Constraints
- Single Job: Only one research job at a time
- Cost Awareness: Always estimate before running
- Memory Persistence: Hypotheses survive across sessions
- QA Gate: Reports below score 70 trigger warnings
Source
git clone https://github.com/blisspixel/primr/blob/main/skills/company-research/SKILL.mdView on GitHub Overview
This skill runs structured company research using Primr to analyze a company or URL. It surfaces prior hypotheses, estimates cost and duration before starting, starts the research job, monitors progress, and finally presents a report with new insights.
How This Skill Works
It begins by checking prior research with get_hypotheses(company). It then estimates the run with estimate_run(company, url, mode) and awaits user approval before starting. Once approved, it launches research_company and periodically polls check_jobs() every 2 minutes until completion, then presents the results and any new hypotheses.
When to Use It
- User asks to research a specific company or analyze its strategy
- User provides a company URL and wants a full investigation
- Competitive analysis or benchmarking against a firm
- Due-diligence or market-entry research for a target company
- Follow-up after initial findings to validate or refine claims
Quick Start
- Step 1: get_hypotheses("Company Name")
- Step 2: estimate_run("Company Name", "https://company.url", "full")
- Step 3: If approved, research_company("Company Name", "https://company.url", "full") and poll check_jobs() until complete; then review results
Best Practices
- Check for prior hypotheses with get_hypotheses(company) before starting
- Always run estimate_run before launching to show cost and duration
- Obtain explicit user approval after presenting the estimate
- Maintain a single active research job per company (no parallel runs)
- Use save_hypothesis to record validated or invalidated claims with evidence
Example Use Cases
- Research Acme Corp at https://acme.com
- Analyze Stellar Foods' business model from https://stellarfoods.co
- Investigate Nova Telecom using https://novatelecom.io
- Evaluate Arcadia Stores at https://arcadiastore.com
- Compare FinNova from https://finova.co
Frequently Asked Questions
Add this skill to your agents